Macrofaunal community structure in the western Indian continental margin including the oxygen minimum zone


Baban S. Ingole, National Institute of Oceanography (Council of Scientific & Industrial Research), Dona Paula, Goa, 403004, India.E-mail:


Patterns of macrofaunal distribution were studied along the western Indian continental margin to distinguish the role of habitat heterogeneity in generating and maintaining community structure. A transect perpendicular to the coast at 14°N latitude was selected for seabed sampling. Eight stations were sampled in the depth range 34–2546 m and characterized with respect to macrofaunal composition, abundance, biomass, diversity and feeding type. The sediments in the shelf region (34, 48, 100 m) and upper slope (525 m) were characterized by silty and sandy facies, whereas the mid slope (1001 m), lower slope (1524 m) and basin (2001, 2546 m) consisted of clayey silts. The highest value of sediment chlorophyll-a (Chl-a) and total organic carbon (Corg) were recorded from the mid slope areas. Faunal abundance and biomass increased from the shallow to deeper depths in the shelf region, and decreased in the slope region (525–1001 m) due to the reduced bottom-water oxygen. The community parameters showed an overall increase in both the lower slope and basin areas. A total of 81 macro-invertebrate species belonging to five major groups represented the macrofauna of the area. Polychaeta was the major group at all depths. Among polychaete families, species of the Spionidae, particularly Prionospio pinnata, predominated at the oxygen minimum zone (OMZ) core and Cossuridae dominated in the lower part of the OMZ in sediments of the mid slope region (1001 m depth). Species diversity was higher in the basin than in the slope region. Fluctuations in diversity appear to be partly due to the bottom-water dissolved oxygen (DO) gradient which includes values that are below the oxygen tolerance of many benthic species. Further, Margalef’s index (d) and Shannon–Wiener index (H′) showed a significant negative (P < 0.01) relationship between sediment Chl-a and Corg, suggesting food availability as a critical factor in species dominance. Results of multivariate analyses suggest that for continental margin fauna, different physiographic provinces and an oxygen gradient have a higher influence on the species composition and diversity than other oceanographic conditions.


In the past few decades, considerable attention has been given to the study of continental margin biodiversity (Flach & Thomsen 1998; Tselepides et al. 2000; Palma et al. 2005). The continental margin, extending from the sublittoral to the abyssal zone, has various interesting habitats that can be described by geomorphological features (e.g. shelf, slope, rise, marginal highs, etc.) and their related environmental conditions (e.g. depth, pressure, temperature, salinity, light, dissolved oxygen, sediment characters and other biogeochemical features). All of these features play an important role in generating and maintaining biodiversity along the continental margin.

The western Indian continental margin, located in the eastern Arabian Sea of the northern Indian Ocean, represents a series of complex environments including a permanent, oxygen depleted zone. Although the Arabian Sea covers only 2% of the surface area of the World Ocean, it is one of the most biologically productive regions (Ryther & Menzel 1965), mainly due to the upwelling of nutrients during the summer, southwest monsoon and convective mixing during the winter, northeast monsoon (Madhupratap et al. 1996; Wiggert et al. 2005). The high biological productivity combines with slow re-oxygenation to produce one of the most intense oxygen depletion zones observed anywhere in the open ocean (Swallow 1984; Naik & Naqvi 2002). This zone intercepts the continental margin at the shelf and continental slope (i.e. bathyal depths), creating extensive seafloor habitats subject to these extreme conditions, which have persisted for hundreds of thousands of years (Reichart et al.1998). These oxygen-depleted zones are known as oxygen minimum zones (OMZs: defined by oxygen concentration <0.5 ml·l−1). The OMZ in the Arabian Sea is spread over ∼285,000 km2 of the benthic area. About 25% of the OMZ has an oxygen concentration <0.5 ml·l−1 and 30% of the area has an oxygen level of <0.2 ml·l−1 (Helly & Levin 2004). This OMZ is the thickest found anywhere in the world and accounts for 40% of the global pelagic N2 production (Bange et al. 2005). According to recently published data, widespread open-ocean oxygen deficiency in this region results from the combination of a high oxygen demand arising from high biological productivity in the surface water and the limited supply of oxygen in intermediate waters (Jayakumar et al. 2009). This low concentration of dissolved oxygen (DO) has a major impact on biogeochemical processes such as the carbon and nitrogen cycles (Naqvi et al. 2006) and on benthic ecological functioning (Levin 2003).

The vertical distribution of benthic populations and their community structure is greatly influenced by the presence of OMZs (Rosenberg et al. 1983; Arntz et al. 1991; Levin 2003). The structure of macrofaunal communities in OMZs typically shows reduced diversity and high dominance in comparison with non-OMZ slope environments (Levin et al. 2001), although the patterns of faunal abundance are less consistent. Biomass is often reduced where oxygen levels are lowest because the macrofauna of the OMZs are generally dominated by small-bodied polychaetes, features likely associated with the low DO, the high availability of food and the reduction in predation pressure (Levin et al. 2002). Previous studies of open-ocean OMZ benthos have suggested a strong lower-boundary effect, with high densities of hypoxic-tolerant faunas aggregating in the lower part of the OMZ (Thompson et al. 1985; Levin 2003; Hughes et al. 2009).

Various geomorphological features occur on the western Indian continental margin. The shelf break in this region occurs between 80 and 110 m, is wider in the northern shelf and narrows progressively towards the south. Various physical, chemical and geological processes control the sedimentation in this region. The sedimentary input is primarily from rivers that drain from the Western Ghats mountain range. Information on the benthos of the deep eastern Arabian Sea, especially from the Indian margin, however, is not available for global comparison. Neyman (1969) studied the benthos of the eastern Arabian Sea and suggested that bottom fauna is poor between 80 and 150 m depth due to an inflow of subsurface water with low oxygen content. Earlier studies on the Indian shelf showed a definite correlation between macrofaunal standing stocks and organic carbon as well as the nature of the substrata (Parulekar & Wagh 1975; Harkantra et al. 1980). The community structure and abundance of nearshore and shallow-water macrofauna is reasonably well known (Jayaraj et al. 2007, 2008a,b; Ingole et al. 2009; Joydas & Damodaran 2009). Most of these reports highlight the influence of environmental factors on the structure of the macrobenthic community. However, most studies were based on sampling at shallow (<200 m) depths in the shelf region. As a result, information on the benthos of the western Indian slope including the OMZ and abyssal zone is not available. Recently, benthic biological and biogeochemical process within the OMZ were investigated on the Pakistan Margin of the northern Arabian Sea (Cowie & Levin 2009; Gooday et al. 2009; Hughes et al. 2009; Levin et al. 2009; Murty et al. 2009).

The aims of this paper are: (i) to generate regional biodiversity data that can be globally compared to other continental margins, (ii) to study the sources of habitat heterogeneity from shallow- to deep-water regions, and (iii) to identify the environmental factors responsible for changing the macrofaunal community structure on the western Indian margin.

It has been recognized that the environmental conditions of the eastern side are different from those acting on the western region of the Arabian Sea (Qasim 1982). Furthermore, the macrofaunal data show a striking contrast (especially through the OMZ core) between the western Oman margin and the eastern Pakistan margin of the northern Arabian Sea (Hughes et al. 2009). In recent studies, increased productivity has been reported from the Arabian Sea as well as the west coast of India (Madhupratap et al. 1996; Prasanna Kumar et al. 2000).

Thus, on the basis of previous studies, we hypothesize that:

  • 1 Increased productivity on the Indian margin will be reflected in standing stocks that are higher than those on the Pakistan margin.
  • 2 Macrofaunal abundance and biomass will be high and diversity will be relatively low in the OMZ region and low DO concentration will be responsible for changing the community structure at some depths.

Oceanographic Settings of the Study Area

The surface area of the Arabian Sea, between latitudes 0° and 25°N and longitudes 50° and 80°E, is about 6.225 × l06 km2. It is bound on the northern, eastern and western sides by the land masses of Asia and Africa. It is an area of negative water balance, where evaporation exceeds precipitation and runoff. The excess of evaporation over precipitation is highest (100–150 cm) off the Arabian coast and decreases steadily towards the southeast. A slight excess of precipitation over evaporation (<20 cm) occurs annually off the southwest coast of India (Venkateswaran 1956).

Surface circulation in the Arabian Sea is controlled by the seasonal variation in winds. Two types of winds blow from different directions and form two different monsoon seasons: the SW monsoon during summer (which leads to precipitation over the entire Indian peninsula) and the NE monsoon during winter. During the SW monsoon, biological productivity in the Arabian Sea lies mainly around the centers of seasonal upwelling off Arabian Peninsula, Somalia and southwest India (Qasim 1977). During this time, the upwelled waters on the southwestern margin of India are restricted by a thin (5–10 m) lens of low-salinity water, which originates from local precipitation and runoff from the narrow coastal plain that receives heavy rainfall (Stramma et al.1996).

Study area

A detailed benthic sampling programme was performed on board ORV Sagar Kanya during August 2007 (cruise no. SK 237) in order to generate local-scale benthic data. A single transect was selected perpendicular to the coast at 14°N latitude (Fig. 1). In this region, the shelf break is located about 105 km from the coast (≈120 m depth), followed by a 116 km wide shelf margin basin. A mid-shelf basement ridge, the dominant visible feature, divides the basin into eastern and western sections. The bottom topography in the mid-lower slope region is strewn with prominent flat-topped marginal highs (Rao & Veerayya 2000; Chakraborty et al. 2006).

Figure 1.

 Location map of the study area.

Material and Methods

Sampling depth and sub-sampling

Eight stations (Table 1) were sampled at water depths of 34, 48, 102, 525, 1001, 1524, 2001 and 2546 m using a spade box corer (50 × 50 × 50 cm size). Two box cores were used to collect sediment samples at each depth, except at station 47 where only one box core was obtained. According to Rao & Veerayya (2000) the transect can be divided into five regions – shelf region (34, 48 and 102 m), upper slope (525 m), mid slope (1001 m), lower slope (1524 m) and basin (2001–2546 m). In the present study, all three slope depths were considered as a single region and therefore the study area was divided into three depth zones – shelf, slope and basin. Sub-sampling was done with a PVC core (15 cm dia) and 1–3 subcores were taken from each box core. Separate sub-cores were collected for organic carbon (Corg), sediment chlorophyll-a (Chl-a) measurements and granulometric texture (grain sizes) analysis and frozen at −20 °C. The water overlying the box core and sub-core sediment samples was sieved through a 300-μm mesh screen and then fixed and preserved using buffered 10% formalin to which rose Bengal was added. Bottom-water salinity and temperature data were collected using a CTD deployed down to 1524 m. CTD data from deeper depths could not be obtained due to technical problems. Bottom-water DO measurements were taken using a DO (dissolved oxygen) sensor attached to the CTD for depths down to 1524 m. Below this depth, water was collected using Niskin bottles and was used for DO and Chl-a analysis.

Table 1.   Geographical position and other information regarding sediment sampling.
Collection dateStationsLat (°N)Long (°E)Depth (m)Number of box coresSub core from box core ASub core from box core B

Laboratory analysis and data processing

DO was analyzed by Winkler’s method (Strickland & Parsons 1968). Chl-a was estimated using an acetone extraction method (Holm-Hansen 1978). Total carbon analyses were carried out on the upper 2 cm of freeze-dried sediments using a NCS 2500 (Model-EA/NA1110) CNS analyzer. Inorganic carbon was analyzed by a CO2 Culometer analyzer and the percentage of CaCO3 was calculated. Percentage of Corg was calculated by subtracting inorganic from total carbon. In the top 2 cm, sediment texture was determined by Malvern Laser Analyzer (Model – Hydro 2000MU).

In the laboratory, the faunal samples were washed on a 300 μm sieve, sorted, identified and counted. Specimens were identified to the lowest possible taxon. Biomass was determined by using the wet weight method after blotting. At each site, the faunal counts from individual subcores (1–6 subcores from one, usually two, box cores) were averaged and the mean value converted to individuals m−2. The faunal counts from the water overlying the box core were divided by the number of subcores taken. The biomass (shell included) was estimated similarly and converted to g·m−2 (wet weight).

The data were subjected to univariate analyses to study the benthic community structure using Margalef’s index (Margalef 1968) for species richness (d), Pielou’s index (Pielou 1966) for species evenness (J′), and the Shannon–Wiener index (Shannon & Weaver 1963) for species diversity (H′ by using loge). The significance of the regions outlined a priori was tested with multivariate analysis [Non-metric Multi-Dimensional Scaling (MDS)] and the organisms that contributed most to the observed differences among groups were found by means of SIMPER (similarity percentage) using PRIMER 6 (Clarke & Warwick 1994). Linear regression between macrofaunal diversity indices and environmental variables was tested using STATISTICA 6. Feeding types were assigned to Polychaeta based on information in Fauchald & Jumars (1979).


Environmental variables

The physico-chemical characteristics of the study area are summarized in Table 2.

Table 2.   Physico-chemical characteristics of habitats examined on the western Indian continental margin.
  1. *N value was very low.

Depth (m)34481025251001152420012546
Bottom water
 Temperature, °C23.220.841810.77.855
 Salinity, psu35.835.535.135.335.235
 DO, ml·l−10.690.560.380.080.281.352.32.3
 Chl aμg·l−
Sediment (0−2)
 Chl aμg·g−
 Corg (%)
 Clay (%)
 Silt (%)75.639.738.553.684.589.479.271.7
Sand (%)16.050.558.533.
 ColourOlive gray
5Y 4/1
Pale yellowish orange
10YR 8/6
Light olive gray
5Y 6/1
Grayish olive green
5GY 3/2
Grayish olive green
10Y 4/2
Yellowish gray
5Y 7/2
Grayish yellow
5Y 8/4
Vary pale orange
10YR 8/2
 Granulometric TextureClayey siltSlity sandSilty sandSandy siltClayey siltClayey siltClayey siltClayey silt

Bottom water

The salinity of the bottom water did not vary at different depths, whereas temperature showed variations between the bathymetric gradients. The bottom-water DO ranged from 0.08 to 2.3 ml·l−1 at different sampling stations. The lowest DO value was on the upper slope and increased from the lower slope reaching a peak in the basin area (Table 2). The highest bottom-water Chl-a was in the shelf region (48 m) and the lowest value was observed on the upper slope. In general, the Chl-a was low at most of the stations (Table 2).


Four types of sediment textures were observed in the study area. The shelf region and upper slopes were characterized by silty and sandy facies, whereas the mid slope, lower slope and Arabian basin were characterized by clayey silts. Sediment Chl-a ranged from 0.2 to 2.1 μg·g−1 and showed significant spatial variation. The middle slope showed the highest Chl-a, and the lowest values were recorded in the shelf (48 m depth) and basin (2546 m) region. Sediment Corg was high on the mid slope (4.4%) and a low value of 0.3% was observed at the basin station (2001 m). Organic carbon was high at stations with high silt content, which was the dominant sediment type in the study area.

Macrofaunal population abundance and biomass

A total of 81 macro-invertebrate species from five major groups represented the macrofauna of the western Indian margin. Of the identified taxa, 67 (82.7%) were Polychaeta, 4 (4.9%) Crustacea, 7 (8.6%) minor Phyla and 3 (3.8%) others (e.g. Bivalvia, Arachnida and Oligochaeta). Polychaeta was by far the dominant macrofaunal group at all the water depths (Fig. 2). The maximum proportion (100%) of polychaetes was observed on the slope (1001 and 1524 m) and the minimum (71.4%) in the basin (2546 m). The next most abundant group was Crustacea with maximum relative abundance (23%) at 2546 m and minimum (3.3%) at 102 m depth.

Figure 2.

 Depth-wise distribution of the macrofaunal groups.

The average macrofaunal abundance increased from depths of 48–102 m depth, then decreased suddenly to a minimum value at the 525-m site, before increasing gradually down to basin depths (2001–2546 m) (Fig. 3). The macrofaunal abundance was much higher at the two deeper stations on the shelf (mean values ∼2907–3722 ind·m−2) than at the shallowest site (mean value 971 ind·m−2). The slope fauna was extremely sparse, with the mean (±SD) abundance of 528 ± 156 ind·m−2. The highest value was recorded on the lower slope and the lowest in the mid slope region. Mean abundance within the basin varied from 1188–1244 ind·m−2.

Figure 3.

 Macrofaunal abundance (ind·m−2) along depth.

The average macrofaunal biomass showed a similar trend. The highest biomass value (13.3 g·m−2) was found at a water depth of 102 m while the lowest value (0.08 g·m−2) was found at 1001 m. Biomass was higher on the upper slope than at other stations (Fig. 4).

Figure 4.

 Macrofaunal biomass (g·m−2) along depth.

Macrofaunal composition

The values for macrofaunal composition in different physiographical provinces are described below and are shown in Table 3.

Table 3.   Taxon list and count data for Western Indian Continental Margin macrofauna (counts are given as number per core, 15 × 10 cm deep).
GroupFamilySpecies nameStations depth (m)
  Arenicola sp.000000000000001000000
  Mediomastus sp.1010100000023101871800000
  Neomediomastus sp.000000000001000100000
  Notomastus sp.000000000000000000000
  Capitellidae sp. 1000000000000000000000
  Cirratulus sp.000000114101000500000
  Cirratulidae sp. 1001000000010962000000
  Cirratulidae sp. 2000000000000000000222
  Cirratulidae sp. 3000000000000000000000
  Tharyx sp.001000000000000000000
  Cossura sp.0000000000071125300105
  Dorvillea sp.0000001301100000000000
  Staurocephalus sp.0000001106130000000000
  Brada sp.000000043000000000000
  Flabelligeridae sp. 1000000000000000000000
  Glycera sp.000000000101000000000
  Goniada sp.000000001000000000000
  Hesione sp.000000000000000000000
  Hesionidae sp. 10000001300300000000000
  Lumbrinereis sp.111010003011100100000
  Lumbrinereis sp. 1000000000000000000000
  Lumbrinereis sp. 2000000000000000000000
  Magelona sp.000000100004000100000
  Magelonidae sp. 1000000000000000001000
  Axiothella sp.000000000000000000000
  Maldanidae sp. 1000000000000000000001
  Nephtys sp.000000001200020000000
  Nereis sp.000000000020000000000
  Onuphis emerita000000010002000010000
  Onuphis sp.000000000000000000000
  Polyophthalmus sp.000000000000000000000
  Opheliidae sp. 1000000000000000000000
  Scoloplos sp.000000000101000000000
  Paraonidae sp. 1000000000000000000000
  Aparaonis sp.000000000000000000000
  Aricidea sp.225211241021036785800000
  Levinsenia sp. 10000009643702252200200
  Levinsenia sp. 2000000000000000000000
  Phyllodoce sp.000000000000000000000
  Ancistrosyllis constricta000000059006365000000
  Ancistrosyllis sp.000100000200000200100
  Pisione sp.000000900000000000000
  Poecilochaetus sp.000000100100000000000
  Chone sp.000000000000000000000
  Jasmineira sp.000000000003341010000
  Sabella sp.000000000000000000000
  Sabellidae sp. 1000000000000000000000
  Sthenelais sp.000000000000000000000
  Prionospio pinnata1326209158122954842540
  Prionospio polybranchiata000000000000000100000
  Prionospio sp.0000000000074107300000
  Spiophanes sp.000000000000000000000
  Spionidae sp. 1000000000000000000000
  Sternaspis sp.300200002020000000000
  Syllis sp.000000037060001000000
  Syllidae sp. 1000000100101000000000
  Syllidae sp. 2000000000000000000000
  Terebellidae sp. 1210000000100322100000
  Terebellidae sp. 2000000000000000000000
 Polychaeta sp. 1 000000000100000000000
 Polychaeta sp. 2 000000000000000000000
 Polychaeta sp. 3 000000000000000000000
 Polychaeta sp. 4 000000000000000000000
 Polychaeta sp. 5 000000000000000000000
 Polychaeta sp. 6 000000000000000000000
 Polychaeta sp. 7 000000000000000000000
 Polychaeta sp. 8 000000000000000000000
  Decopoda Natantia010010010001000000000
Bivalvia  000000002000000000000
Minor Phyla                       
  Dendrostomus sp.000000000000000000000
  Phascolion sp.000000000000000000000
  Sipuncula sp. 1000100000010000000000
  Sipuncula sp. 2000000000000000000000
  Sipuncula sp. 3000000000000000000000
Oligochaete  021007252220100000000
GroupFamilySpecies name1524-A21524-B12001-A12001-A22001-A32001-B12001-B22001-B32546-A12546-A22546-A32546-B12546-B22546-B3
  Arenicola sp.00000000000000
  Mediomastus sp.00000000000000
  Neomediomastus sp.00000000000000
  Notomastus sp.00000000100000
  Capitellidae sp. 101100000000000
  Cirratulus sp.00000000502010
  Cirratulidae sp. 100000000000000
  Cirratulidae sp. 221000000000000
  Cirratulidae sp. 300221010120301
  Tharyx sp.00000000000000
  Cossura sp.20000000000000
  Dorvillea sp.00000000000000
  Staurocephalus sp.00000000000000
  Brada sp.10010001000000
  Flabelligeridae sp. 112000000000000
  Glycera sp.21000010000000
  Goniada sp.00100011021131
  Hesione sp.00000000010000
  Hesionidae sp. 100002000000000
  Lumbrinereis sp.00001000000000
  Lumbrinereis sp. 100100011032001
  Lumbrinereis sp. 200000000000010
  Magelona sp.00000000000100
  Magelonidae sp. 100000000000000
  Axiothella sp.00000000002000
  Maldanidae sp. 101000000000000
  Nephtys sp.00000000000000
  Nereis sp.00010010020011
  Onuphis emerita00000000000000
  Onuphis sp.00000000000010
  Polyophthalmus sp.00000001000000
  Opheliidae sp. 100000010000000
  Scoloplos sp.00000000000000
  Paraonidae sp. 100044000000000
  Aparaonis sp.00000000000001
  Aricidea sp.01020010010000
  Levinsenia sp. 100000000000000
  Levinsenia sp. 243410000100000
  Phylodoce sp.00200002000000
  Ancistrosyllis constricta00000000000000
  Ancistrosyllis sp.00000000000100
  Pisione sp.00000000000000
  Poecilochaetus sp.02000000000000
  Chone sp.00000010000000
  Jasmineira sp.00000000000000
  Sabella sp.00000000100000
  Sabellidae sp. 100010010210000
  Sthenelais sp.00000000100010
  Prionospio pinnata00000000000100
  Prionospio polybranchiata00000000000000
  Prionospio sp.00100020245312
  Spiophanes sp.00001010000000
  Spionidae sp. 100055000500000
  Sternaspis sp.00000001002110
  Syllis sp.00010000010000
  Syllidae sp. 100000000000000
  Syllidae sp. 200010000201031
  Terebellidae sp. 100000000000000
  Terebellidae sp. 201001000000121
 Polychaeta sp. 1 00000000000000
 Polychaeta sp. 2 00031000000000
 Polychaeta sp. 3 00041000000000
 Polychaeta sp. 4 00004000000000
 Polychaeta sp. 5 00002000000000
 Polychaeta sp. 6 00050000000000
 Polychaeta sp. 7 00000200100000
 Polychaeta sp. 8 00020000000000
  Decopoda Natantia00000000000000
Bivalvia  00030000110010
Minor Phyla                
  Dendrostomus sp.0000004000000 
  Phascolion sp.00001000000000
  Sipuncula sp. 100001011001000
  Sipuncula sp. 200002000000000
  Sipuncula sp. 30002 000000000
Oligochaete  00000000000000

Shelf (34, 48 and 102 m)

A total of 43 taxa were identified and Polychaeta was the dominant group, contributing 85–95% to the faunal abundance. Of the 28 polychaete families identified in the entire area, 24 were observed on the shelf. The family Paraonidae made the highest contribution (36.8%) followed by Spionidae (16.9%) and Capitellidae (9.3%). Decopoda Natantia and Oligochaeta were present while Amphipoda were not recorded. Other groups such as Bivalvia and minor Phyla were found in low numbers.

Slope (525, 1001 and 1524 m)

Only two groups, Polychaeta and Crustacea, were present at this depth range. The majority (80–100%) of the macrofaunal animals were polychaetes, which were represented by 14 families. The contribution of Cossuridae was maximal (62.5%) at the mid slope area. Spionidae were predominant on the upper slope. The maximum number of families (10) was observed in the lower slope area. The only crustacean taxon present in this region was Amphipoda.

Basin (2001 and 2546 m)

The highest number of taxa (52) was present in the basin region. This area was marked by a lower proportion (69–71%) of Polychaeta than the other two regions. A total of 21 polychaete families was observed, with the Spionidae and Paraonidae being dominant (24.5% and 14% respectively). The highest contribution of Crustacea (23% at 2546 m) and minor Phyla (12.1% at 2001 m) were also observed in this region. Tanaidacea was the most abundant group among the crustaceans, contributing ∼12% to the total macrofaunal community.

Dominant taxa

The three dominant species (representing 22–100% of the macrofaunal community) at their respective depths, together with their rank, are summarised in Table 4 and Fig. 5.

Table 4.   Top three ranked taxa observed at different water depths (rank based on the species abundance) (‘+’ indicates the species was present but not ranked).
TaxonWater depth (m)
Aricidea sp.1+3  +++
Prionospio pinnata2221   +
Sternaspis sp.3+    ++
Levinsenia sp. 1 1+     
Staurocephalus sp. 3      
Mediomastus sp.+ 1     
Cossura sp.  ++12  
Amphipoda   3  ++
Cirratulidae sp. 2   22+  
Maldanidae sp. 1    3+ +
Paraonidae sp. 1   +  3 
Levinsenia sp. 2     1++
Flabelligeridae sp. 1+    +  
Tanaidacea ++   11
Spionidae sp. 1      2+
Prionospio sp. ++   +2
Figure 5.

 Rank 1 species at various depths and their abundance pattern along the sampling transect.

Among these species, the polychaetes Aricidea sp., Prionospio pinnata and Sternaspis sp. were dominant at a depth of 34 m. However, their faunal contribution decreased sharply as water depth increased. At a depth of 48 m, the three top-ranked species were the polychaetes Levinsenia sp. 1, P. pinnata and Staurocephalus sp.; again their abundances declined with increasing water depth. Mediomastus sp. was dominant at 102 m, but was not observed at any other depth except for at the shallowest site (34 m), where Prionospio pinnata was ranked second dominant.

Prionospio pinnata, Cirratulidae sp. 2 and an unidentified Amphipoda species dominated in the upper slope area. Only three species were observed in the mid slope region, with Cossura sp. dominating. Levinsenia sp. 2 was present on the lower slope. The polychaete Cirratulidae sp. 2 was the only taxon restricted to the slope region.

The highest taxonomic diversity was observed in the basin. Here, Tanaidacea were dominant at both depths (2001–2546 m). Tanaidacea were also present on the shelf, but with low abundance and were absent on the slope. At 2001 m, Spionidae sp. 1 and Paraonidae sp. 1 appeared as second and third ranked, respectively. The polychaete Prionospio sp. occupied second rank at the 2546 m depth.

Multivariate (MDS) analysis of community structure

An MDS plot based on the average abundance of macrofauna revealed three distinct groups at 20% similarity, reflecting the three different physiographic regions (Fig. 6) in the study area. Group 1 was restricted to the shelf region and was characterized by the dominance of P. pinnata, A. constricta and Aricidea sp. Group 2 covered the slope region (mid and lower) and was dominated by Cossura sp. and Cirratulidae sp. 2. Group 3 consisted of the two deeper stations located in the Arabian basin, with polychaetes Levinsenia sp. 2 and Cirratulidae sp. 3, along with Tanaidacea, as the dominant taxa. The upper slope remained separate as it did not cluster with any of the above groups.

Figure 6.

 MDS analysis of macrofaunal community.

Diversity indices

Margalef’s index (d) varied from 0.3 to 5.5 over the study area (Fig. 7). A higher d value was recorded from the basin (2001 m) than in the mid slope. Species evenness varied from 0.6 to 0.9, with the higher value recorded at a depth of 2001 m, rather than at a shallower depth (34 m) on the shelf. Values of H′ varied from 0.9 to 3.4. The highest H′ value was also observed in the basin (2001 m).

Figure 7.

 Depth-wise distribution of community structure indices (d: Margalef index, J′: evenness and H′: Shannon diversity).

Distribution of polychaete feeding types

Surface deposit-feeders (SDF) were the dominant feeding type at all the sampling depths (Fig. 8). The highest contribution of SDF was on the slope. The frequency of SDF was also high (50–75%) within the slope area. The mid slope showed the highest contribution (50%) of sub-surface deposit feeders (SSDF). The highest representation of carnivores was on the shelf region (particularly at 48 m). Moreover, a moderately high contribution of carnivores was observed at the two basin sites. Carnivores were not present in the mid slope area and filter feeders were rare across the study area.

Figure 8.

 Depth-wise distribution of macrobenthic feeding types (%) (SDF, surface deposit feeders; SSDF, sub-surface deposit feeders; C, carnivores; FF, filter feeders).

Correlation of environmental and community parameters

The correlation between macrobenthic parameters and environmental variables were based on Pearson’s correlation analyses (Table 5). Macrofaunal abundance was positively correlated with sand and CaCO3 (P < 0.05). The H′ values were negatively correlated with sediment Chl-a and Corg (P < 0.05). Margalef’s index showed a significant negative relationship between Corg and silt content (P < 0.05) and a positive relationship with sand and CaCO3 (P < 0.05). Only species evenness showed a significant negative relationship with bottom water salinity (P < 0.05). Other environmental variables, including DO, did not show significant correlations with macrofaunal community parameters.

Table 5.   Linear regression analyzed for eight environmental variables against measures of macrofaunal community parameters at 8 stations along the study area.
  1. n.s. = Not significant.

Depth (m)n.s.n.s.n.s.n.s.n.s.
Salinity (psu)n.s.n.s.n.s.r = −0.83, P = 0.037n.s.
Chl an.s.n.s.n.s.n.s.n.s.
Sed Chln.s.n.s.n.s.n.s.r = −0.91, P = 0.01
TOCn.s.n.s.r = −0.87, P = 0.02 r=−0.83, P = 0.04
Clay (%)n.s.n.s.n.s.n.s.n.s.
Silt (%)n.s.n.s.r = −0.81, P = 0.05 n.s.
Sand (%)r = 0.85, P = 0.02n.s.r = 0.84, P = 0.03 n.s.
CaCO3r = 0.86, P = 0.02n.s.r = 0.87, P = 0.02 n.s.


The observed high values of sediment Corg on the slope, and low values in the basin region, were in agreement with previous studies of the western Indian margin. According to Rao & Veerayya (2000), diverse topographic features on the slope and the associated hydrodynamic processes play an important role in the enrichment of Corg. DO values ranged from 0.08 to 2.3 ml·l−1. The lowest DO value was measured at the upper slope station. Based on DO concentrations, the OMZ was found to extend from a water depth of 102–1001 m in the study area, with the core of the OMZ located at 525 m, where the lowest DO value was recorded.

Macrofaunal biomass

Macrofaunal biomass increased from 34 to 102 m, then showed an abrupt decline with the lowest value measured at 1001 m, before increasing again from 1001 m to 2546 m. Previous studies from the shelf region (Harkantra et al. 1980; Parulekar et al. 1982; Harkantra 2004) report higher biomass values at the shelf stations. The differences noted may be due to the varied sampling methods including different types of gear, water depth and the sampling season. Moreover, all the previous studies sampled at shallower depths (<200 m). Although the biomass in the present study was low, the values increased on the shelf from the shallower to the deeper regions, a pattern not reported in earlier studies (Kurian 1971; Parulekar & Dwivedi 1974; Ansari et al. 1977; Jayaraj et al. 2007, 2008b).

The high benthic biomass at shallower depths in the shelf region may reflect higher food availability in the form of Corg and Chl-a. The supply of food to the sub-tidal benthic environment depends on the proximity to the shore and water depth (Levinton 1982). Sediment Chl-a at 34 m and 102 m was moderately high. Furthermore, biomass was higher at 102 m compared to shallower regions, due to the presence of Decapoda Natantia and Arenicola sp. Although the highest Corg was recorded in the lower part (1001 m) of the OMZ, biomass was lower at this depth and was related to faunal abundance. Comparatively high biomass was observed at 525 m, which was not reflected in abundance. This was due to the presence of the large-sized species P. pinnata. There was an increase in biomass thereafter to the lower boundary of the OMZ towards the deeper region.

Macrofaunal abundance and composition

Macrofaunal abundance increased from the shallow to the deeper region of the shelf, decreased to its lowest value in the lower part of the OMZ and then increased towards the deeper region. Macrofaunal abundance in the western Indian OMZ core (525 m) and the lower part of the OMZ were extremely low compared to all other OMZ margins studied to date, except for the Pakistan margin located on the northeastern side of the Arabian Sea (Table 6). In the present study, low faunal abundance was observed on the slope at depths where DO concentrations were 0.08–0.28 ml·l−1, although measurements of Chl-a and Corg were high. Levin et al. (2009) suggested that oxygen thresholds for macrofaunal abundance exist at ∼0.11–0.13 ml−1 when the other conditions are favorable. It is possible that oxygen thresholds in our study area are higher than on the Pakistan margin.

Table 6.   Comparison of macrofaunal abundance (ind·m−2) in the western and eastern continental margin of the Arabian Sea.
Water depth (m)DO (ml·l−1)Western India (present study)Pakistan (Hughes et al. 2009)Pakistan (Levin et al. 2009)Oman (Levin et al. 2000)
1402.12 14,894  
 0.11 10,464  
3000.11 323  
4000.13   12,362
7000.16  016,283
7530.119  414 
8030.122  1656 
 0.137  414 
8500.2   19,193
8530.125  2622 
 0.14  414 
9030.128  2967 
 0.147  6762 
9400.13 5222  
 0.17 3380  
9530.134  4544 
 0.156  2898 
 0.27   5818
10030.146  3726 
 0.174  4692 
10530.164  2208 
 0.199  690 
12000.35 1003  
12500.52   2485
18501.72 8531  

Among the polychaetes, Spionidae were highly dominant within the OMZ core, whereas Cirratulidae were the predominant taxon in other regions. Spionidae and Cirratulidae were also the dominant families within the OMZ on the upper slope (400–700 m) off Oman (Levin et al. 2000) and at a shallow shelf station (140 m), the lower OMZ boundary (1200 m) and below the OMZ (1850 m) on the Pakistan margin (Hughes et al. 2009). In the present study, cossurids were mainly restricted to the lower part of the OMZ (1001 m), although they were also present on the lower slope which is located just beneath the OMZ. This polychaete group was abundant in the OMZ core (100–200 m) off Central Chile (Gallardo et al. 2004). Because these polychaetes are deposit-feeders (both surface and sub-surface), their predominance may reflect food availability (e.g. sediment Corg and Chl-a) within the OMZ region. Paraonidae were abundant at sites above and below the OMZ regions as well as the upper shelf and the basin. Similarly, paranoids were abundant at 1850 m on the Pakistan margin (Hughes et al. 2009).

The Cossuridae and Spionidae were important taxa where oxygen concentrations were lowest. Cossuridae are common in many bathyal OMZs, including the Pakistan margin where they are present at depths of 940–1200 m, both areas of high Corg. Furthermore, the global bathyal data on benthic faunal abundance and biomass indicate a reduction in density at the OMZ core (Rosenberg et al. 1983; Mullins et al. 1985; Wishner et al. 1995) and an increase at the OMZ boundaries (Levin 2003).

Influence of habitat heterogeneity on macrofaunal community structure

The MDS ordination based on the macrofaunal community clustered the sites into three groups representing three different bathymetric provinces. Group 1 comprised stations of the shelf region, where opportunistic species such as Aricidea sp., Prionospio pinnata, Mediomastus sp. and Ancistrosyllis constricta were dominant. These polychaete species were most abundant at a depth of 102 m, where the DO concentration was low (<0.5 ml·l−1). For most of these species, this tolerance of stressful conditions such as the low DO (<0.5 ml·l−1) has been observed in the OMZs of the Oman margin and off central Chile (Levin et al. 2000; Gallardo et al. 2004).

Group 2 was restricted to the slope region (1001–1524 m), where the SSDF species Cossura sp. and Cirratulidae sp. 2 predominated within the macrofaunal community. Although 1524 m was outside the OMZ, it was close to the OMZ boundary and the measured DO was not as high, nor as stable as the values measured at deeper stations.

Group 3, the most diverse group, was dominated by different taxa, including Tanaidacea, Levinsenia sp. 2 and Cirratulidae sp. 2. This group was confined to the two deepest (basin) sites, of which the 2546 m region had a higher value of Corg (excluding those measured at the OMZ stations). The nature and variability of the organic matter supplied to the deep-sea floor influence the structure and function of the communities (Grassle & Morse-Porteous 1987). Tanaidacea are not very diverse in shallow waters but are well represented in the deep sea (Dojiri & Sieg 1997; Pavithran et al. 2007), where they are one of the most abundant taxa. Among the crustaceans, Isopoda and Amphipoda are also particularly abundant and diverse in the deep sea, where they are among the most typical members of the benthic communities (Sanders et al. 1965; Brandt et al. 2007). The diversity of the Group 3 assemblage in the present study was comparable to that of ‘normal’ deep-sea habitats.

The OMZ core fauna did not cluster with any group. This was due to the higher abundance of P. pinnata (Fig. 5), which is known to tolerate low oxygen concentrations (Gallardo et al. 2004). The dominance of P. pinnata has also been reported within the OMZ off Concepecion (Palma et al. 2005). In addition, one particular morphological adaptation of this species, an expanded branchial structure, has been observed only in the OMZ settings, specifically at the lowest level of oxygen concentration on the upper slope (Fig. 9) and not in any other habitat included in the present study. Some other species were restricted to specific areas. Mediomastus sp., Arenicola sp., Levinsenia sp. 1, Ancistrosyllis constricta were confined to the shelf, Cirratulidae sp. 2 to the slope and Lumbrinereis sp. 2, Polyopthalmus sp., Cirratulidae sp. 3 to the basin sites (Table 3). Of the non-polychaete taxa, oligochaetes were found only on the shelf and thus were also restricted to the upper portion of the OMZ. Oligochaet have been reported from the low-oxygen zone in a basin off Peru with partially laminated sediments (Levin et al. 2002). During the present study, macrofaunal diversity was generally higher at shelf and basin sites than on the slope, reflecting the lower concentration of oxygen in the slope region.

Figure 9.

 Photograph of anterior part of Prionospio pinnata with well-developed branchia observed in the upper slope OMZ.

The bathymetric distribution of polychaete feeding types is directly related to the amount of organic matter available in the sediments, with SDF strongly dominating the slope regions. Moreover, the proportion of carnivores in the shelf and basin regions increased with depth, proportional to the decline in the deposit-feeding component. The positive relationship between Corg and deposit feeders, and the increase of carnivores where deposit feeders were less abundant, has been observed at the Oman and Crete margins (Levin et al. 2000; Tselepides et al. 2000).

The result of the present study suggests that macrobenthic community structure on the western Indian margin is not determined by a single factor, but instead is influenced by a combination of environmental factors. Discussing animal-sediment relationships, Snelgrove & Butman (1994) concluded that the complexity of soft-sediment communities may defy any simple paradigm with regard to any single factor controlling their settlement and colonization. The distribution of annelids within OMZs worldwide has been reviewed by Levin (2003), who has suggested the different patterns of community structure are due to changes of hydrodynamic, bathymetric or geochemical factors rather than dissolved oxygen alone. Among the biological parameters, abundance and biomass positively correlated with sand and correlated negatively with silt. This is because the sand percentage was higher in shallow water, where higher faunal abundance and biomass were also higher. Similarly, P. pinnata, Cossura sp. and Ancistrosyllis sp. were also observed in sandy sediment at low oxygen concentrations on the western Indian shelf (Jayaraj et al. 2008b). The increase of Chl-a, in both water and sediment, was related to the enhanced phytoplankton production in the study area.

Summary and Conclusions

This study reveals several novel characteristics of macrofaunal communities and their response to habitat heterogeneity on the western Indian margin. The physiographic provinces and their related environmental characteristics in the study area generated habitat heterogeneity, which is summarized below together with the corresponding community characteristics.

  • 1 The shelf (34, 48, 102 m) was dominated by sandy sediment with low DO. It included part of the OMZ at 102 m, and had moderately high sediment Corg content. The shelf contained the highest abundance and biomass with moderately high diversity, species richness and SDF feeding types.
  • 2 The slope (525, 1001, 1524 m) was characterized by silty sediment and included the OMZ above the lower slope with higher Corg. Diversity and species richness were lowest and the percentage of SDF feeding types was highest here.
  • 3 The basin (2001, 2546 m) appears to be a normoxic region with silty texture and lower Corg content. This region displayed the highest diversity, species richness and presence of fauna with the maximum numbers of feeding types.

Dominant taxa and faunal composition differed along the gradient of habitat heterogeneity. In general, results from the Pakistan margin were weak predictors for macrofaunal community structure along the Indian margin. The reason for low abundance and biomass in the core and lower boundary parts of the OMZ in the western Indian margin compared to other areas is not clear. Furthermore, the results of the present study did not support our second hypothesis, as macrofaunal abundance and biomass were lower in the OMZ region, except in the shallowest part (40, 102 m) where abundance and biomass were high. We believe that further investigation based on seasonal sampling in the shelf region and high-resolution sampling in the OMZ region is required to understand the community interaction with seasonal, environmental changes on the western Indian margin.


The authors are thankful to Dr S.R. Shetye, Director, NIO (Goa, India) for his encouragement. We also express our gratitude to the scientific team and crew members of ORV Sagar Kanya for their help during sampling. We wish to acknowledge the support received from Dr M. Sudhakar of MoES for ship time. Comments and suggestions from three anonymous reviewers and Prof. Lisa Levin helped in improving the manuscript. Our special thanks to Ms. Jennifer Gonzalez, Scripps Institution of Oceanography, La Jolla, California and Andy Gooday, National Oceanography Centre, UK for meticulously going through the manuscript and suggesting language corrections. This is the contribution no. 4657 of NIO (CSIR) Goa.